能谱CT鉴别卵巢原发良恶性肿瘤的初步研究

发布时间：2018-03-04 00:33

  本文选题：卵巢肿瘤　切入点：宝石能谱CT　出处：《山东大学》2017年硕士论文　论文类型：学位论文

【摘要】：目的:通过对符合研究条件的卵巢原发良恶性肿瘤患者进行能谱CT扫描,获得单能量能谱曲线图、碘含量、水含量并计算能谱曲线斜率等参数,能谱CT诊断结果与病理结果对比分析,初步探讨能谱CT在卵巢原发良恶性肿瘤鉴别诊断中的临床应用价值。方法:回顾性分析2015年5月至2016年8月进行盆腔动脉、静脉期能谱CT扫描的女性患者58人,年龄范围19～64岁,将其分为两组:良性组30例,恶性组28例。选取标准:①检查者为女性;②病变均含有实性成分;③无对比剂过敏;④具有本院或其他三级医院病理学检查结果。所有患者均在宝石能谱CT(Discovery CT 750 HDGE Heahhcare,Mliwaukee,美国)上进行盆腔平扫加双期增强扫描,扫描范围至少包括髂脊到耻骨联合下缘水平。双期增强扫描均用能谱扫描模式(GemStone imaging,GSI),管电压在80kVp、140kVp间0.5ms内迅速切换,管电流:375mAs,转速:0.5s/周,探测器宽:40mm,扫描展厚及层间距均:5.00mm。高压注射器将碘海醇(300mgl/mL)通过肘静脉注入,剂量:1.2mL/kg,流速::3.0 mL/s。动脉期图像于注射开始后25-30s扫描得到,静脉期图像于注射开始后60-75s扫描得到。采用标准算法重建图像。所有患者扫描前均签署知情同意书。将获得的能谱图像拆为层厚和层间距1.25mm单能量图像,传入AW4.4工作站进行能谱分析,采用能谱成像软件(GSI Viewer)进行单能量图像显示、定量数据测量。在肿瘤较多实性成分的层面放置感兴趣区(ROI),RO]放置标准:①选取实性成分最大层面放置相同大小的圆形或椭圆形ROI,再选取其上、下两个层面测量,取三者平均值作为测量结果;②ROI面积为50～60mm2;③测量位置尽量在两期相同的层面;④放置ROI时避开囊性部分、瘤体坏死、液化、钙化及血管区。然后测算R0I40～100keV单能量下(间隔10keV)平均CT值、碘含量(ICg/L)、水含量(WCg/L)、40～100keV间能谱曲线斜率并获得ROI能谱曲线,40～100keV间能谱曲线斜率λ=CT(40keV)-CT(100keV)/|40keV-100keV|)。将两组分析结果对比相应的病理结果。计量(定量)经正态性检验和方差齐性检验,满足正态性的再根据方差齐性的结果采用t检验或校正t检验,不满足正态性的采用秩和检验。计数(定性)资料的比较分析采用χ2检验,Kappa值用于评价影像与病理结果的一致性。所有数据均应用SAS9.4统计软件分析,P0.05为有统计学意义。结果:两组病变在动脉期、静脉期随着单能量增大能谱曲线形态呈弧形下降型,能量越低CT值差异越大;动脉期单能量40～100keV间、静脉期60～140keV间两组CT值差异均有统计学意义(P0.05),动脉期良性组CT值大部分小于恶性组,静脉期良性组CT值大部分大于恶性组;动脉期40～1OOkeV间能谱曲线斜率、碘含量良性组分别为(1.43±0.58)、(11.82±4.89)100g/L,恶性组分别为(2.43±0.29)、(20.77±1.90)100g/L,两组差异均有统计学意义(P0.05);动脉期两组水含量两组分别为(1030.37± 10.56)、(1025.96±6.10)100g/L,静脉期两组水含量分别为(1032.34±11.44)、(1029.06±6.25)100g/L,静脉期碘含量、能谱曲线斜率良性组分别为(17.78±5.75)100g/L、(1.98±0.65),恶性组分别为(15.85±2.02)100g/L、(2.0±0.38),差异均无统计学意义(P0.05);能谱CT对良性、恶性肿瘤的检出率分别为86.67%(26/30)、85.71%(24/28),灵敏度分别为 86.67%、80.00%,特异度分别为 80.00%、86.67%,配对设计χ2检验显示,能谱CT与病理结果无统计学差异(χ2=30.39,P0.0001)Kappa值为0.72,一致性较高。结论及意义:能谱CT动脉期单能量40～100keV间CT值、动脉期能谱曲线斜率及动脉期碘含量良性组均小于恶性组,能谱CT诊断结果与病理结果无统计学差异(Kappa值为0.72),一致性较高。说明能谱CT多参数分析对卵巢良恶性肿瘤的诊断具有一定的参考价值及临床应用意义。
[Abstract]:Objective: to meet the conditions on the primary ovarian benign and malignant tumor spectrum CT scan were obtained, the energy spectrum curve, iodine content, water content and calculation of energy spectrum curve slope parameters compared to the spectral analysis of the CT diagnosis and pathological results, preliminary study of spectral CT in the diagnosis of primary ovarian identification of benign and malignant tumors in clinical application. Methods: a retrospective analysis from May 2015 to August 2016 for pelvic artery, venous phase spectrum CT scans of 58 female patients, age ranged from 19~64 years old, they were divided into two groups: 30 cases of benign group and malignant group 28 cases. Selection criteria: check for female; the lesions contain solid components; and without contrast agent allergy; the hospital has three hospitals or other pathological results. All patients were in the gemstone CT (Discovery CT 750 HDGE Heahhcare, Mliwaukee, America) on the pelvic plain and Dual phase enhanced scanning, the scanning range includes at least the iliac crest to the lower edge of pubic symphysis. Dual phase enhanced scan with spectrum scan mode (GemStone imaging, GSI), tube voltage in 80kVp, 140kVp 0.5ms quickly switch tube current: 375mAs speed: 0.5s/ weeks, detector width: 40mm, scanning exhibition the thickness and spacing of layers are: 5.00mm. high pressure syringe will iohexol (300mgl/mL) through elbow vein injection dose: 1.2mL/kg, flow rate: 3 mL/s. arterial phase images in 25-30s scanning after injection, intravenous injection in the image after the start of 60-75s scan. The standard algorithm for image reconstruction. All patients were scanned before signed informed consent will be obtained. Energy spectrum image split with layer thickness of 1.25mm single energy image was transferred to AW4.4 workstation, energy spectrum analysis, using spectral imaging software (GSI Viewer) single energy image display, quantitative data in swollen measurements. The tumor more solid component level placement of region of interest (ROI), RO] put standard: round or oval ROI the solid component placed the largest level of the same size, then select it, under the two level measurement, the three is taken as the average value of the measurement results; the ROI area is 50 ~ 60mm2; 3 to measure the position of the same level as far as possible in the two periods; avoid the cystic part of the placement of ROI, tumor necrosis, liquefaction, calcification and vascular area. Then measure the R0I40 ~ 100keV single energy (10keV interval) the average value of CT, the content of iodine (ICg/L), water content (WCg/L), 40 ~ 100keV spectrum curve the slope and ROI spectrum curve, 40 ~ 100keV, energy spectrum curve slope of lambda =CT (40keV) -CT (100keV) /|40keV-100keV|). The pathological results of two group analysis results are compared. The corresponding measurement (Quantitative) by the test of normality and homogeneity of variance test, to satisfy the normality again according to variance the results of the T test or calibration t test, Wilcoxon test does not meet the normality. Count (Qualitative) data analysis using 2 test, Kappa value for consistency evaluation of imaging and pathological results. All data are used SAS9.4 statistical analysis software, P0.05 was considered statistically significant. Results: two groups of lesions in the arterial phase, venous phase with single energy increases energy spectrum curve shape curved down type, the lower the energy difference of the CT value is higher; the arterial phase single energy 40 ~ 100keV, 60 ~ 140keV in venous phase two groups were statistically significant differences in CT values (P0.05), arterial CT value is less than most of malignant and benign group group, benign group venous phase CT value greater than the malignant group; arterial phase 40 ~ 1OOkeV spectrum curve slope, iodine content in benign group respectively (1.43 + 0.58), (11.82 + 4.89) 100g/L, malignant group were (2.43 + 0.29), (20.77 + 1.90) 100g/L, there were two groups the difference 鎰忎箟(P0.05);鍔ㄨ剦鏈熶袱缁勬按鍚�閲忎袱缁勫垎鍒�涓，

本文编号：1563397